1. What is in scope
An edge gateway, an industrial server, and a 5G MEC node are treated on the SimtoLink Device Management Platform as a resource you can orchestrate, with a life close to that of a cloud service. It starts with secure bootstrapping, goes onto the asset register, receives network settings and a resource quota, then stream rules and containers, and stays under health monitoring and patches until retirement and a secure wipe. The node record has to point at its sub-device mapping, so a retirement or a swap has a plan for moving the workload and handing over the data.
Discovery and approval
Find the node from a certificate or a pre-shared key. A person approves it before it joins a production group, so unknown hardware does not come online.
GitOps for configuration
Push one DSL and one rule version, and keep them in two-way sync with the cloud Git repository. A high-risk change needs two people to review it.
Rolling upgrade and rollback
Use a canary and a rolling update. If the health check fails, roll back to the last known-good version.
The life of a node
| Capability | What it does |
|---|---|
| Registration and discovery | Discover from a certificate or a pre-shared key, then group by tag after approval |
| Policy sync | DSL and configuration sync both ways through GitOps. What the console shows is what the repository holds |
| Canary upgrade | A rolling update is tied to the health check. On failure, roll back and send a webhook |
| What you can see | Export metrics to Prometheus. Logs and diagnostic bundles follow the offline buffer and the bandwidth policy |
| Retirement | Secure wipe, revoke the certificate, reclaim API credentials, and line up with audit and CMDB |
2. Discovery, approval, and groups
A node can be discovered from a certificate or a pre-shared key, but it does not enter a production group until a person approves it. That keeps unknown hardware out. Tags record the room, the rack position, the maintenance contract, and the power circuit, which is what a technician needs on site.
3. Configuration, rules, and GitOps
Push one DSL and one rule version, and keep a two-way sync with the cloud Git repository, so what the console shows is what the repository holds. A high-risk change needs two people to review it.
4. Upgrade, canary, and rollback
Edge software uses a canary and a rolling update. If the health check fails, it rolls back to the last known-good version and pushes an event on a webhook.
5. Metrics and logs
Export metrics to Prometheus. Logs and diagnostic bundles come back under the offline buffer and the bandwidth policy, so they do not crowd out control traffic.
6. High availability
A critical node should run gateway high availability, or a cold-standby switchover playbook, and the platform should record which role it holds now.
7. Retirement and compliance
Wipe the node, revoke the certificate, reclaim API credentials, and remove it from the CMDB. The steps follow lifecycle management and the audit requirement.
8. Summary
An edge node is not a one-time install. It is a long-lived asset that should be governed from the same source as the cloud. SimtoLink closes the loop from bootstrap to retirement. Further reading: gateway high-availability failover and edge function deployment. For a reference architecture for edge GitOps, contact the SimtoLink delivery team.
Appendix. Operating the node fleet after go-live
An edge lifecycle that looks finished in a pilot still needs three routines before it covers a fleet. On a pre-release batch that matches production, rehearse a connection storm, a disk that fills, certificate rotation, and a fault that crosses regions, and file the result as a checklist you can run again. Write the health metrics, alert thresholds, on-call path, and audit retention into change management, so a quota or a GitOps rule can be traced and rolled back without depending on whoever last touched the node. Once a quarter, engineering, operations, security, and data should check whether those signals still explain the risk, and tie the follow-ups to tickets and the release cadence. Console, open API, and export on the SimtoLink Device Management Platform are assets you keep running: observable, orchestrable, and auditable. If the design has to meet your ITSM, IAM, data lake, or edge cluster, the SimtoLink solutions team can review it with you and return an implementation note and a risk list.